Key takeaways
- S&P has published the SPIVA scorecard since 2002, comparing active funds to the benchmark they chose for themselves.
- Over one year the results are noisy. Over fifteen, the majority of active managers have trailed in every equity category.
- The scorecard corrects for survivorship, which matters because a large share of funds do not last the full period.
- Persistence is the deeper problem. Past top performers repeat at rates no better than chance.
What SPIVA is
Twice a year, S&P Dow Jones Indices publishes a scorecard comparing actively managed funds against the index each fund holds itself to. It has run since 2002.
Two design choices make it worth reading. It uses the fund's own stated benchmark rather than a convenient one, and it corrects for survivorship by counting funds that closed or merged during the period. That second point matters more than it sounds, because funds that close are rarely the ones that were winning.
The pattern gets clearer with time
Over a single year the split looks like something close to a coin toss with costs attached. Extend the window and the picture stops being ambiguous. For the fifteen years ending in December 2024, there was not a single equity category in which a majority of active managers beat their benchmark. Not one out of twenty two.
The shares shown above are indicative of the long running pattern rather than a fixed figure. The direction has been consistent for two decades.
Why it happens
Not because managers are unintelligent. Because of arithmetic and costs.
Before fees, active investors as a group must earn the market return, since collectively they are the market. After fees they must earn less. The average active fund starts each year behind by roughly its expense ratio and has to make that back through skill before it produces anything.
The persistence problem
The natural response is to pick the funds that have been winning. S&P publishes a separate persistence scorecard for exactly that, and the result is unhelpful. Top quartile funds repeat at rates that are no better than random, and often worse.
If past performance identified skill, persistence would show up. It does not, which suggests most of what looks like skill over short windows was variance.
Sources
- S&P Dow Jones Indices, SPIVA US Scorecard, published since 2002 using the survivorship bias free CRSP mutual fund database.
- For the fifteen years ending December 2024, SPIVA reported no equity category in which a majority of active managers outperformed.
- S&P Dow Jones Indices, SPIVA Persistence Scorecard, for the finding that top quartile performance does not persist above chance.
- Look up the current scorecard for present figures. The pattern is durable, any single year is not.
Open Forecast. Run this math on your own numbers, with every assumption labeled.
Open in Capital Allocation →